Fix all 114 broken documentation links

docs/engineering/ADAPTER_AUTHOR_GUIDE.md

@@ -38,7 +38,7 @@ Adapters may not:
 
 ## Required adapter card
 
-Every adapter should have a model/adapter card using [`MODEL_CARD_TEMPLATE.md`](MODEL_CARD_TEMPLATE.md).
+Every adapter should have a model/adapter card using [`MODEL_CARD_TEMPLATE.md`](../trust/MODEL_CARD_TEMPLATE.md).
 
 The card must include:
 

docs/engineering/ARCHITECTURE.md

@@ -92,19 +92,19 @@ The purpose is to improve which questions qualified humans choose to test next.
 | `openamp_foundry.active_learning` | Informative next-batch selection under uncertainty. |
 | `openamp_foundry.analysis` | Diversity, similarity, structural-class, and audit helpers. |
 
-Package behavior must remain aligned with docs under [`docs/PROJECT_INDEX.md`](PROJECT_INDEX.md).
+Package behavior must remain aligned with docs under [`docs/PROJECT_INDEX.md`](../PROJECT_INDEX.md).
 
 ## Core artifacts
 
 | Artifact | Purpose | Source doc |
 |---|---|---|
-| Evidence certificate | Explains candidate selection or rejection. | [`EVIDENCE_CERTIFICATE.md`](EVIDENCE_CERTIFICATE.md) |
+| Evidence certificate | Explains candidate selection or rejection. | [`EVIDENCE_CERTIFICATE.md`](../evidence/EVIDENCE_CERTIFICATE.md) |
 | Run manifest | Records command, inputs, config, commit, hashes, and claim boundaries. | [`RUN_MANIFEST_STANDARD.md`](RUN_MANIFEST_STANDARD.md) |
-| Benchmark card | Documents benchmark purpose, data, baselines, and limits. | [`BENCHMARK_GOVERNANCE.md`](BENCHMARK_GOVERNANCE.md) |
-| Dataset card | Documents source, license, labels, preprocessing, bias, and release status. | [`DATA_GOVERNANCE.md`](DATA_GOVERNANCE.md) |
-| Model or adapter card | Documents model/adapter scope, benchmarks, limitations, and release status. | [`MODEL_CARD_TEMPLATE.md`](MODEL_CARD_TEMPLATE.md) |
-| Review packet | Packages artifacts for qualified external review. | [`EXTERNAL_REVIEW_PACKET.md`](review/EXTERNAL_REVIEW_PACKET.md) |
-| Decision record | Records important governance decisions. | [`DECISION_RECORD_TEMPLATE.md`](operations/DECISION_RECORD_TEMPLATE.md) |
+| Benchmark card | Documents benchmark purpose, data, baselines, and limits. | [`BENCHMARK_GOVERNANCE.md`](../evidence/BENCHMARK_GOVERNANCE.md) |
+| Dataset card | Documents source, license, labels, preprocessing, bias, and release status. | [`DATA_GOVERNANCE.md`](../trust/DATA_GOVERNANCE.md) |
+| Model or adapter card | Documents model/adapter scope, benchmarks, limitations, and release status. | [`MODEL_CARD_TEMPLATE.md`](../trust/MODEL_CARD_TEMPLATE.md) |
+| Review packet | Packages artifacts for qualified external review. | [`EXTERNAL_REVIEW_PACKET.md`](../review/EXTERNAL_REVIEW_PACKET.md) |
+| Decision record | Records important governance decisions. | [`DECISION_RECORD_TEMPLATE.md`](../operations/DECISION_RECORD_TEMPLATE.md) |
 
 ## Threat model
 

docs/engineering/README.md

@@ -23,7 +23,7 @@ Current structural convention:
 ## Extension points
 
 - [Adapter author guide](ADAPTER_AUTHOR_GUIDE.md)
-- [Model card template](../MODEL_CARD_TEMPLATE.md)
-- [Virtual assay scope](../VIRTUAL_ASSAY_SCOPE.md)
+- [Model card template](../trust/MODEL_CARD_TEMPLATE.md)
+- [Virtual assay scope](../evidence/VIRTUAL_ASSAY_SCOPE.md)
 
 Return to the [documentation home](../README.md).

docs/engineering/SCHEMA_REGISTRY.md

@@ -20,17 +20,17 @@ Future work should add machine-readable schema metadata and automated schema che
 
 | Artifact type | Purpose | Stability target | Source docs |
 |---|---|---|---|
-| Candidate evidence certificate | Explains why a candidate was selected, rejected, or reviewed. | High. | [`EVIDENCE_CERTIFICATE.md`](EVIDENCE_CERTIFICATE.md) |
-| Candidate manifest | Lists panel candidates, roles, hashes, release status, and proof level. | High. | [`WET_LAB_HANDOFF.md`](review/WET_LAB_HANDOFF.md) |
+| Candidate evidence certificate | Explains why a candidate was selected, rejected, or reviewed. | High. | [`EVIDENCE_CERTIFICATE.md`](../evidence/EVIDENCE_CERTIFICATE.md) |
+| Candidate manifest | Lists panel candidates, roles, hashes, release status, and proof level. | High. | [`WET_LAB_HANDOFF.md`](../review/WET_LAB_HANDOFF.md) |
 | Run manifest | Records command, config, inputs, commit, versions, and hashes. | High. | [`RUN_MANIFEST_STANDARD.md`](RUN_MANIFEST_STANDARD.md) |
-| Benchmark card | Documents benchmark purpose, data, baselines, metrics, caveats, and status. | High. | [`BENCHMARK_GOVERNANCE.md`](BENCHMARK_GOVERNANCE.md) |
-| Dataset card | Documents dataset source, license, labels, preprocessing, bias, and release status. | High. | [`DATA_GOVERNANCE.md`](DATA_GOVERNANCE.md) |
-| Model or adapter card | Documents intended use, not-for list, data, benchmarks, safety, and release status. | High. | [`MODEL_CARD_TEMPLATE.md`](MODEL_CARD_TEMPLATE.md) |
-| External review packet | Bundles candidate, benchmark, safety, or adoption evidence for external review. | Medium-high. | [`EXTERNAL_REVIEW_PACKET.md`](review/EXTERNAL_REVIEW_PACKET.md) |
-| Result summary | Captures qualified outcome summaries at a safe abstraction level. | High. | [`CALIBRATION_POLICY.md`](CALIBRATION_POLICY.md) |
-| Calibration intake report | Joins predictions with structured result summaries. | High. | [`CALIBRATION_POLICY.md`](CALIBRATION_POLICY.md) |
-| Recalibration gate verdict | Records whether recalibration may be considered. | High. | [`CALIBRATION_POLICY.md`](CALIBRATION_POLICY.md) |
-| Decision record | Records important project decisions. | Medium-high. | [`DECISION_RECORD_TEMPLATE.md`](operations/DECISION_RECORD_TEMPLATE.md) |
+| Benchmark card | Documents benchmark purpose, data, baselines, metrics, caveats, and status. | High. | [`BENCHMARK_GOVERNANCE.md`](../evidence/BENCHMARK_GOVERNANCE.md) |
+| Dataset card | Documents dataset source, license, labels, preprocessing, bias, and release status. | High. | [`DATA_GOVERNANCE.md`](../trust/DATA_GOVERNANCE.md) |
+| Model or adapter card | Documents intended use, not-for list, data, benchmarks, safety, and release status. | High. | [`MODEL_CARD_TEMPLATE.md`](../trust/MODEL_CARD_TEMPLATE.md) |
+| External review packet | Bundles candidate, benchmark, safety, or adoption evidence for external review. | Medium-high. | [`EXTERNAL_REVIEW_PACKET.md`](../review/EXTERNAL_REVIEW_PACKET.md) |
+| Result summary | Captures qualified outcome summaries at a safe abstraction level. | High. | [`CALIBRATION_POLICY.md`](../evidence/CALIBRATION_POLICY.md) |
+| Calibration intake report | Joins predictions with structured result summaries. | High. | [`CALIBRATION_POLICY.md`](../evidence/CALIBRATION_POLICY.md) |
+| Recalibration gate verdict | Records whether recalibration may be considered. | High. | [`CALIBRATION_POLICY.md`](../evidence/CALIBRATION_POLICY.md) |
+| Decision record | Records important project decisions. | Medium-high. | [`DECISION_RECORD_TEMPLATE.md`](../operations/DECISION_RECORD_TEMPLATE.md) |
 
 ## Required schema metadata
 

docs/evidence/NOVELTY_AUDIT_GUIDE.md

@@ -12,7 +12,7 @@ Every AMP candidate must pass a full novelty audit before wet-lab synthesis to:
 2. **Ensure scientific validity** — a "novel" label on a KNOWN_VARIANT of a published AMP is misleading.
 3. **Protect the budget** — synthesis of a patent-risk or known candidate wastes money.
 
-The single biggest mistake in OpenAMP Wave 0.5 was relying on the `novelty-check-broad` CLI tool, which uses a 72-sequence curated reference — far too small to catch patent proximity. See [`docs/postmortems/2026-06-29-seed020-novelty-patent-risk.md`](postmortems/2026-06-29-seed020-novelty-patent-risk.md) for the full incident.
+The single biggest mistake in OpenAMP Wave 0.5 was relying on the `novelty-check-broad` CLI tool, which uses a 72-sequence curated reference — far too small to catch patent proximity. See [`docs/postmortems/2026-06-29-seed020-novelty-patent-risk.md`](../postmortems/2026-06-29-seed020-novelty-patent-risk.md) for the full incident.
 
 **Rule: Always run `scripts/novelty/run_expanded_novelty_audit.py` before adding any candidate to the panel or ordering synthesis.**
 

docs/getting-started/AGENT_ONBOARDING.md

@@ -19,10 +19,10 @@ The lab is the judge.
 Before changing anything, read:
 
 1. [`../AGENTS.md`](../AGENTS.md) — primary operating contract.
-2. [`../SAFETY.md`](../SAFETY.md) — safety policy.
-3. [`../RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md) — allowed and disallowed use.
-4. [`../MISSION.md`](../MISSION.md) — scientific boundaries.
-5. [`PROJECT_INDEX.md`](PROJECT_INDEX.md) — navigation map.
+2. [`../SAFETY.md`](../) — safety policy.
+3. [`../RESPONSIBLE_USE.md`](../) — allowed and disallowed use.
+4. [`../MISSION.md`](../) — scientific boundaries.
+5. [`PROJECT_INDEX.md`](../PROJECT_INDEX.md) — navigation map.
 6. [`METRICS_CURRENT.md`](../evidence/METRICS_CURRENT.md) — current evidence and known weaknesses.
 7. [`DECISION_RULES.md`](../evidence/DECISION_RULES.md) — gates and thresholds.
 

docs/getting-started/FIRST_RUN_WALKTHROUGH.md

@@ -15,9 +15,9 @@ A good first run should produce clarity, not confidence.
 Read:
 
 - [`README.md`](../README.md)
-- [`SAFETY.md`](../SAFETY.md)
+- [`SAFETY.md`](../)
 - [`docs/getting-started/COMMAND_SURFACE.md`](COMMAND_SURFACE.md)
-- [`docs/evidence/PROOF_LADDER.md`](PROOF_LADDER.md)
+- [`docs/evidence/PROOF_LADDER.md`](../evidence/PROOF_LADDER.md)
 
 ## Expected setup
 

docs/getting-started/HUMAN_ONBOARDING.md

@@ -27,7 +27,7 @@ If you want to make dramatic claims faster than the evidence supports, this is t
 ## First 30 minutes
 
 1. Read [`README.md`](../README.md).
-2. Read [`SAFETY.md`](../SAFETY.md).
+2. Read [`SAFETY.md`](../).
 3. Run:
 
 ```bash
@@ -48,36 +48,36 @@ Read the docs that match your role.
 ### Engineer path
 
 - [`docs/engineering/ARCHITECTURE.md`](../engineering/ARCHITECTURE.md)
-- [`CONTRIBUTING.md`](../CONTRIBUTING.md)
+- [`CONTRIBUTING.md`](../)
 - [`docs/evidence/BENCHMARKING.md`](../evidence/BENCHMARKING.md)
 - [`docs/evidence/DECISION_RULES.md`](../evidence/DECISION_RULES.md)
 
 Then look for a small issue in CLI ergonomics, tests, reports, schemas, or doc consistency.
 
 ### Scientist path
 
-- [`VISION.md`](../VISION.md)
-- [`GOAL.md`](../GOAL.md)
-- [`docs/evidence/PROOF_LADDER.md`](PROOF_LADDER.md)
+- [`VISION.md`](../)
+- [`GOAL.md`](../)
+- [`docs/evidence/PROOF_LADDER.md`](../evidence/PROOF_LADDER.md)
 - [`docs/evidence/METRICS_CURRENT.md`](../evidence/METRICS_CURRENT.md)
 - [`docs/evidence/BENCHMARKING.md`](../evidence/BENCHMARKING.md)
 
 Then look for a weak benchmark, missing baseline, leakage risk, uncontrolled comparison, or overclaim.
 
 ### Lab-review path
 
-- [`docs/review/WET_LAB_HANDOFF.md`](WET_LAB_HANDOFF.md)
-- [`docs/review/COLLABORATION_PLAYBOOK.md`](COLLABORATION_PLAYBOOK.md)
-- [`docs/evidence/PROOF_LADDER.md`](PROOF_LADDER.md)
-- [`RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md)
+- [`docs/review/WET_LAB_HANDOFF.md`](../review/WET_LAB_HANDOFF.md)
+- [`docs/review/COLLABORATION_PLAYBOOK.md`](../review/COLLABORATION_PLAYBOOK.md)
+- [`docs/evidence/PROOF_LADDER.md`](../evidence/PROOF_LADDER.md)
+- [`RESPONSIBLE_USE.md`](../)
 
 Then review whether the evidence package would help a qualified scientist decide whether a small assay batch is worth considering.
 
 ### Safety path
 
-- [`SAFETY.md`](../SAFETY.md)
-- [`RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md)
-- [`MODEL_RELEASE_POLICY.md`](../MODEL_RELEASE_POLICY.md)
+- [`SAFETY.md`](../)
+- [`RESPONSIBLE_USE.md`](../)
+- [`MODEL_RELEASE_POLICY.md`](../)
 - [`docs/evidence/DECISION_RULES.md`](../evidence/DECISION_RULES.md)
 
 Then look for release risk, overclaiming, unsafe defaults, unclear boundaries, or missing human-review gates.

docs/getting-started/MAINTAINER_GUIDE.md

@@ -182,7 +182,7 @@ Positive results are more dangerous than negative results because they invite ov
 
 Before any public claim:
 
-1. Map the result to [`PROOF_LADDER.md`](PROOF_LADDER.md).
+1. Map the result to [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md).
 2. Check novelty carefully.
 3. Check safety-adjusted interpretation.
 4. Compare against baselines.

docs/getting-started/README.md

@@ -5,8 +5,8 @@
 1. [Repository README](../../README.md)
 2. [First-run walkthrough](FIRST_RUN_WALKTHROUGH.md)
 3. [Command surface](COMMAND_SURFACE.md)
-4. [Methods](../METHODS.md)
-5. [Current metrics](../METRICS_CURRENT.md)
+4. [Methods](../engineering/METHODS.md)
+5. [Current metrics](../evidence/METRICS_CURRENT.md)
 
 ## Choose a role
 

docs/operations/HIGH_LEVERAGE_TASKS.md

@@ -53,7 +53,7 @@ Upgrade: stronger claim discipline.
 Verification:
 
 - no Level 6+ language for dry-lab artifacts;
-- claims map to [`PROOF_LADDER.md`](PROOF_LADDER.md);
+- claims map to [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md);
 - forbidden phrases removed or qualified.
 
 ### Task: Candidate release audit

docs/research/WAVE_0_5_EXTERNAL_PREDICTOR_SUMMARY.md

@@ -20,7 +20,7 @@ historical process context, not as the current result state.
 
 The external screen was completed later and the authoritative current-state summary
 now lives in [`METRICS_CURRENT.md`](../evidence/METRICS_CURRENT.md)
-and [`ROADMAP.md`](research/ROADMAP.md).
+and [`ROADMAP.md`](ROADMAP.md).
 
 ---
 

docs/review/ASSAY_PREREGISTRATION.md

@@ -25,8 +25,8 @@ The pilot does not ask whether any candidate is a drug, therapy, cure, clinical
 - [`PRE_REGISTERED_PILOT_TEMPLATE.md`](PRE_REGISTERED_PILOT_TEMPLATE.md) — general pilot template.
 - [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md) — review packet standard.
 - [`WET_LAB_HANDOFF.md`](WET_LAB_HANDOFF.md) — safe expert-review handoff guide.
-- [`PROOF_LADDER.md`](PROOF_LADDER.md) — evidence levels and claim ladder.
-- [`CALIBRATION_POLICY.md`](CALIBRATION_POLICY.md) — gate before any recalibration.
+- [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md) — evidence levels and claim ladder.
+- [`CALIBRATION_POLICY.md`](../evidence/CALIBRATION_POLICY.md) — gate before any recalibration.
 - [`NEGATIVE_RESULT_ARCHIVE.md`](../evidence/NEGATIVE_RESULT_ARCHIVE.md) — safe negative-result format.
 
 ## Pilot identity
@@ -156,7 +156,7 @@ After structured result intake:
 6. Require human review before applying any change.
 7. Record the decision in a dated decision log.
 
-See [`CALIBRATION_POLICY.md`](CALIBRATION_POLICY.md).
+See [`CALIBRATION_POLICY.md`](../evidence/CALIBRATION_POLICY.md).
 
 ## Claim boundaries
 
@@ -182,7 +182,7 @@ Forbidden:
 
 ### After results
 
-Claims must follow [`PROOF_LADDER.md`](PROOF_LADDER.md).
+Claims must follow [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md).
 
 Examples:
 

docs/review/COLLABORATION_PLAYBOOK.md

@@ -205,7 +205,7 @@ Without baseline comparison, a positive outcome is hard to interpret.
 
 Responsible result publication should include:
 
-- exact claim level from [`PROOF_LADDER.md`](PROOF_LADDER.md);
+- exact claim level from [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md);
 - candidate evidence package where safe;
 - selection rule;
 - baseline comparison;

docs/review/EXPERT_REVIEW_PACK.md

@@ -47,7 +47,7 @@ Forbidden language before sufficient evidence:
 - proven;
 - AI-discovered antibiotic.
 
-Use [`PROOF_LADDER.md`](PROOF_LADDER.md) for evidence levels.
+Use [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md) for evidence levels.
 
 ## Required packet structure
 
@@ -278,9 +278,9 @@ Do not imply that a reviewer endorsed claims outside their review scope.
 - [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md) — broader packet standard.
 - [`WET_LAB_HANDOFF.md`](WET_LAB_HANDOFF.md) — safe expert-review handoff guide.
 - [`PRE_REGISTERED_PILOT_TEMPLATE.md`](PRE_REGISTERED_PILOT_TEMPLATE.md) — pilot planning template.
-- [`PROOF_LADDER.md`](PROOF_LADDER.md) — claim levels.
+- [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md) — claim levels.
 - [`BENCHMARK_GOVERNANCE.md`](../evidence/BENCHMARK_GOVERNANCE.md) — benchmark lifecycle.
-- [`MODEL_RELEASE_POLICY.md`](../MODEL_RELEASE_POLICY.md) — release boundaries.
+- [`MODEL_RELEASE_POLICY.md`](../) — release boundaries.
 
 ## Final standard
 

docs/review/LAB_PARTNER_ONBOARDING.md

@@ -105,7 +105,7 @@ A panel summary may include:
 - known model blind spots;
 - release status.
 
-Full candidate identities or sequences should be released only according to [`MODEL_RELEASE_POLICY.md`](../MODEL_RELEASE_POLICY.md), [`RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md), and safety review.
+Full candidate identities or sequences should be released only according to [`MODEL_RELEASE_POLICY.md`](../), [`RESPONSIBLE_USE.md`](../), and safety review.
 
 ## Partner evaluation questions
 
@@ -176,7 +176,7 @@ Operational experimental details should not be added to this repository.
 
 ## Publication and claim boundaries
 
-Before any public result is described, map it to [`PROOF_LADDER.md`](PROOF_LADDER.md).
+Before any public result is described, map it to [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md).
 
 Dry-lab candidate package:
 
@@ -203,9 +203,9 @@ Not allowed without much stronger evidence:
 
 All partner-facing work should follow:
 
-- [`SAFETY.md`](../SAFETY.md)
-- [`RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md)
-- [`MODEL_RELEASE_POLICY.md`](../MODEL_RELEASE_POLICY.md)
+- [`SAFETY.md`](../)
+- [`RESPONSIBLE_USE.md`](../)
+- [`MODEL_RELEASE_POLICY.md`](../)
 - [`COLLABORATION_PLAYBOOK.md`](COLLABORATION_PLAYBOOK.md)
 - [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md)
 
@@ -242,14 +242,14 @@ Suggested labels:
 - `wet-lab-facing`
 - `do-not-automerge`
 
-See [`ISSUE_LABEL_TAXONOMY.md`](ISSUE_LABEL_TAXONOMY.md).
+See [`ISSUE_LABEL_TAXONOMY.md`](../operations/ISSUE_LABEL_TAXONOMY.md).
 
 ## Related documents
 
 - [`WET_LAB_HANDOFF.md`](WET_LAB_HANDOFF.md) — safe expert-review handoff guide.
 - [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md) — review packet standard.
 - [`PRE_REGISTERED_PILOT_TEMPLATE.md`](PRE_REGISTERED_PILOT_TEMPLATE.md) — non-protocol pilot planning template.
-- [`PROOF_LADDER.md`](PROOF_LADDER.md) — evidence and claim ladder.
+- [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md) — evidence and claim ladder.
 - [`COLLABORATION_PLAYBOOK.md`](COLLABORATION_PLAYBOOK.md) — safe collaboration modes.
 - `schemas/lab_result.schema.json` — machine-readable result-summary schema.
 

docs/review/PRE_REGISTERED_PILOT_TEMPLATE.md

@@ -43,7 +43,7 @@ It does not ask whether any candidate is a drug, treatment, cure, clinical candi
 
 ## Evidence level before testing
 
-Before testing, every candidate must be treated as Level 4 or below under [`PROOF_LADDER.md`](PROOF_LADDER.md): a computationally nominated candidate with a reviewable evidence package.
+Before testing, every candidate must be treated as Level 4 or below under [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md): a computationally nominated candidate with a reviewable evidence package.
 
 ## Candidate panel
 
@@ -141,7 +141,7 @@ Forbidden:
 - AI-discovered antibiotic;
 - proven.
 
-After results, claims must follow [`PROOF_LADDER.md`](PROOF_LADDER.md).
+After results, claims must follow [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md).
 
 ## Safety review
 

docs/review/WET_LAB_HANDOFF.md

@@ -2,7 +2,7 @@
 
 **Status:** Safe expert-review handoff, not a protocol.
 **Audience:** Qualified domain experts, safety reviewers, and potential institutional partners.
-**Source-of-truth companions:** [`PROOF_LADDER.md`](PROOF_LADDER.md), [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md), [`PRE_REGISTERED_PILOT_TEMPLATE.md`](PRE_REGISTERED_PILOT_TEMPLATE.md), [`RESPONSIBLE_USE.md`](../RESPONSIBLE_USE.md).
+**Source-of-truth companions:** [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md), [`EXTERNAL_REVIEW_PACKET.md`](EXTERNAL_REVIEW_PACKET.md), [`PRE_REGISTERED_PILOT_TEMPLATE.md`](PRE_REGISTERED_PILOT_TEMPLATE.md), [`RESPONSIBLE_USE.md`](../).
 
 ## Purpose
 
@@ -110,7 +110,7 @@ Required:
 - claim-level update;
 - next-batch recommendation or rejection.
 
-Claim level: depends on evidence; see [`PROOF_LADDER.md`](PROOF_LADDER.md).
+Claim level: depends on evidence; see [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md).
 
 ## Score interpretation for reviewers
 

docs/trust/PUBLICATION_POLICY.md

@@ -41,10 +41,10 @@ Before public claims, complete:
 
 Use:
 
-- [`PROOF_LADDER.md`](PROOF_LADDER.md)
+- [`PROOF_LADDER.md`](../evidence/PROOF_LADDER.md)
 - [`CLAIM_REVIEW_CHECKLIST.md`](../evidence/CLAIM_REVIEW_CHECKLIST.md)
 - [`RELEASE_CHECKLIST.md`](RELEASE_CHECKLIST.md)
-- [`SAFETY.md`](../SAFETY.md)
+- [`SAFETY.md`](../)
 
 ## Allowed dry-lab language